Introduction

Coronary artery ectasia (CAE) has been defined as the inappropriate dilatation of a coronary artery, with the luminal diameter 1.5 or more times wider than that of adjacent normal segments [1]. Its prevalence has been reported to range from 0.3 to 4.9 % according to previous autopsy and angiography studies [1, 2]. Clinically, it predisposes patients to adverse coronary events, such as vasospasm, thrombosis and dissection [3]. The pathogenesis of CAE remains elusive. More than 50 % of CAE patients have obstructive coronary artery disease (CAD), indicating that CAD could be the main underlying aetiology of CAE [4]. Characteristic changes in CAE are extensive destruction of musculoelastic elements, including damage of elastin fibres and a decreased number of smooth muscle cells [4–6]. Makis noted that CAE patients showed confirmed evidence of cystic medial necrosis [6], and previous studies showed that proteolytic enzymes may play vital roles in the pathogenesis of CAE [7–9].

Normally, the coronary media contains many layers of lamellar units, which are composed of vascular smooth muscle cells lying between two layers of elastin fibres and surrounded by type I and type III collagen and other extracellular matrix (ECM) proteins [10]. Elastin fibres are the dominant ECM proteins in the coronary wall, constituting up to 50 % of the dry weight [11, 12], and supporting the elasticity and tensile strength of the vessels [13]. Due to a lack of de novo synthesis of elastin after reaching adulthood, its degradation is irreversible and can lead to various symptoms and diseases [13]. Increased levels of proteolytic enzymes, including serine, thiol, aspartic and metallo enzymes, can promote fragmentation of the elastin fibre network in the media. The most widely studied enzymes are elastases, such as neutrophil elastase and pancreatic elastase, which belong to the serine family, and matrix metalloproteinases (MMPs) [13]. After destruction, soluble elastin (sElastin) and its fragments are released into the circulating blood. Thus, the degradation of elastin fibres can be detected by evaluating blood sElastin [11].

Type I and type III collagen are the most abundant collagen fibres in the artery wall. Collagen type I consists of thick and strong fibres, and it mainly provides rigidity to the vessels. In contrast, type III collagen is thin and interweaved into a fine reticular network, which gives elasticity to tissue and organs [14, 15]. Various circulating markers can be used to indicate the turnover of collagens. Collagen type I cross-linked telopeptides (ICTP) and the protocollagen III N-terminal propeptide (PIIINP) are cleavage products of degraded type I [16] and type III collagen [17], respectively. Thus, ICTP and PIIINP can be used as collagen degradation makers. The procollagen type I carboxy terminal peptide (PICP) and procollagen a1(III) C-terminal propeptide (PIIICP) are fragments produced during the biosynthesis of type I [18] and type III [19] collagen and act as markers of collagen synthesis. As it is not easy to gather human coronary artery autopsy specimens, levels of ICTP, PIIINP, PICP and PIIICP in serum can be used as an alternative method to assess coronary ECM metabolism in CAE patients.

To the best of our knowledge, the turnover of ECM, including elastin fibres and type I and type III collagens, has not been studied before in patients with CAE. Thus, the present study evaluated the main components of ECM metabolism in CAE patients by measuring levels of circulating sElastin, ICTP, PIIINP, PICP and PIIICP. We also determined the main ECM proteolytic enzyme activities, including total elastase activity and total MMP activity in the same patients [20, 21]. Those results would be useful clues for the further in-depth mechanism researches for CAE.

Patients and methods

Patient population

In total, 1640 patients’ serum samples were collected after coronary angiograms immediately from September 2009 to March 2013 in the cardiac catheterisation centre of Peking Union Hospital. The indications of angiographic examination for these patients were stable angina, acute coronary syndrome and atypical chest discomfort. Among these patients, 51 (3.11 %) of them had CAE and 30 of them without conditions listing in exclusion criteria were included in the CAE group. During the same operation period, we randomly selected 30 patients with angiographically documented CAD and 30 subjects with relatively normal coronary arteries (named Control or CON). The three groups were balanced in age, sex and other baseline characteristics. This study was approved by the local ethics committee and was in accord with the Declaration of Helsinki. Informed consent was obtained from all the study participants.

Inclusion criteria

Coronary angiograms were performed via a femoral approach using the Judkins technique, without the use of nitroglycerin, adenosine or a calcium channel blocker. Each angiogram was interpreted by two independent cardiologists. CAE was defined as an ectatic artery diameter exceeding 1.5 times that of adjacent normal segments [6]. CAD was defined as 50 % or more stenosis in one or more major coronary arteries [9]. Subjects without abnormal coronary arteries were used as controls.

Exclusion criteria

The exclusion criteria were acute coronary syndrome, cardiomyopathy, valvular heart disease, heart failure, collagen tissue diseases, vasculitis, syphilis, chronic obstructive lung disease, pulmonary hypertension, early menopause, organic hepatic diseases, chronic alcoholism, use of steroid or anti-inflammatory drugs within the last 3 months, renal failure and cancer.

Medical records and blood samples

The hospital medical records were detailed and intact. Most of the data in this research were extracted from the medical records. The blood samples were collected immediately after the coronary angiography, and serum samples were separated within 6 h and kept at −80 °C for the future experiments.

ELISA for the detection of circulating sElastin, ICTP, PIIINP, PICP and PIIICP

Quantitative determination of sElastin, ICTP, PIIINP, PICP and PIIICP was performed using sandwich ELISA kits according to the manufacturer’s instructions (Elabscience Biotechnology Co.,Ltd, Wuhan, China). Briefly, the serum samples were first incubated in microtitre wells precoated with anti-sElastin, anti-ICTP, anti-PIIINP, anti-PICP and anti-PIIICP antibodies. The microtitre wells were then washed, and peroxidase-labelled FabV antibody was added to each well. The amount of peroxidase bound to each well was determined by the addition of 3,3 V,5,5 V-tetramethylbenzidine substrate. The reaction was stopped with acid solution, and the resultant colours were read at 450 nm in a microtitre plate spectrophotometer. The final concentrations were calculated by interpolation from a standard curve.

Total elastase activity assay

Total elastase activity was measured using the method employed by Bizbiz [20] and Harris [21]. Briefly, serum samples (50 µL) from different groups were incubated with 150 µL of polypeptide substrate N-succinyl-(l-alanine)3-p-nitroanilide (1 mmol/L, Sigma-Aldrich [Shanghai] Trading Co., Ltd) dissolved in Tris–HCl (200 mmol/L; pH 8.0) for 2 h at 37 °C in a 96-well plate. Then, 5 µL of glacial acetic acid was added to stop the reaction. On cleavage by elastase, this substrate is absorbed at 405 nm. The absorbtion value at 405 nm of each sample was determined and compared to a calibration curve prepared with standard elastase solution and its dilutions (Sigma-Aldrich [Shanghai] Trading Co., Ltd). Elastase activity was expressed as the activity percent compared to that of standard elastase solution.

Total MMP activity assay

The total activity of MMPs was determined using a GenMed MMP Fluorimetric Assay Kit (GenMed Scientifics Inc., Shanghai, China) according to the manufacturer’s instructions. Briefly, 50 µL of serum samples was added to a reaction mixture containing MMP fluorescence resonance energy transfer peptide substrate, Mca-PLGL-Dap-(Dnp)-AR-OH [22]. After 40 min of incubation at 37 °C, the fluorescence intensity was measured at wavelengths of 330 nm/400 nm (excitation/emission) with a fluorescence microplate reader, using 7-methoxycourmarin (MCA) as the standard. One unit of MMP activity was defined as the amount of enzyme that hydrolysed 1 nmol of peptide substrate per min at 37 °C at pH 7.5.

Statistical analysis

Statistical analyses were performed with SPSS 17.00 package software. General descriptive characteristics were assessed as mean ± standard deviation (SD). Nonparametric data characteristics were assessed as percent (%). Sex and risk factors were compared between groups using the Chi square test. For parametric variables, statistical differences among the groups were tested with a one-way analysis of variance or Kruskal–Wallis test. The LSD method or Nemenyi test was used for multiple comparisons between the groups. The correlations between sElastin and total elastase activity and total MMP activity were evaluated with Pearson’s correlation test. The lowest level of significance was accepted as P < 0.05.

Results

The baseline characteristics of the CAE, CAD and Control are given in Table 1. The three groups were similar with regard to age, sex, presence of hypertension, fasting glucose, lipid profile, renal function, and most other cardiovascular risk factors (P > 0.05). The diastolic blood pressure, body mass index, hs-CRP, and blood monocyte number were higher in the CAE and CAD groups. The main results were the following:

Table 1 Clinical character istics of the collected subjects
Full size table
  1. 1.

    Increased circulating sElastin in patients with CAE

As shown in Table 2 and Fig. 1, the level of sElastin was significantly higher in the CAE group than in the CAD (P1 = 0.009) and Control groups (P2 = 0.000). It was also higher in the CAD group than the Control group (P3 = 0.015). The data indicate that the rate of elastin fibre degradation was higher in the CAE group than in the other two groups.

Table 2 Circulating sElastin, ICTP, PIIINP, PICP and PIIICP and activities of proteolytic enzymes
Full size table
Fig. 1
figure 1

Serum sElastin concentration. *P < 0.05, **P < 0.01. CAE coronary artery ectasia, CAD coronary arthrosclerosis disease, CON Control group, sElastin soluble elastin, a marker of elastin fibre degradation

Full size image
  1. 2.

    Turnover of type I collagen in CAE patients

As shown in Table 2 and Fig. 2, there was no significant difference in the serum ICTP concentrations among the three groups (P = 0.168). However, the serum PICP concentration was significantly higher in the CAE group compared to the other two groups (P1 = 0.001, P2 = 0.002). The data reveal that the synthesis, but not the degradation of type I collagen, was greater in the CAE group.

Fig. 2
figure 2

Serum ICTP,PIIINP, PICP and PIIICP concentration. *P < 0.05, **P < 0.01, # P > 0.05. ICTP collagen type I cross-linked telopeptides, a marker of type I collagen degradation; PICP procollagen type I carboxy terminal peptide, a marker of type I collagen synthesis; 2) PIIINP protocollagen III N-terminal propeptide, a marker of type III collagen degradation; PIIICP procollagen a1(III) C-terminal propeptide, a marker of type III collagen synthesis

Full size image
  1. 3.

    Turnover of type III collagen in CAE group.

According to the data in Table 2 and Fig. 2, the concentration of PIIINP increased significantly in the CAE group compared to that of the other two groups (P1 = 0.002, P2 = 0.007). In contrast, there was no statistically significant difference in the concentration of PIIICP among the three groups (P = 0.079). The results demonstrate that the degradation, but not the synthesis of type III collagen, increased in the CAE group.

  1. 4.

    Increased serum proteolytic enzyme activities in the CAE group.

The total elastase activity was significantly higher in the CAE group than in the CAD and Control groups (P1 = 0.006, P2 = 0.022). The data imply that elastase activity was higher in the CAE group than in the other two groups (Table 2; Fig. 3).

Fig. 3
figure 3

Serum total elastase activity and total MMP activity. *P < 0.05, **P < 0.01, # P > 0.05. CAE coronary artery ectasia, CAD coronary arthrosclerosis disease, CON Control group

Full size image

The data showed that the total MMP activity was significantly higher in the CAE group than in the Control group (P2 = 0.013) but not significantly higher than in the CAD group (P1 = 0.477). The data point to a tendency of higher MMP activity in the CAE group (Table 2; Fig. 3).

As shown in Fig. 4, in all selected subjects, correlation coefficients between sElastin and total elastase activity had significant meaning (r = 0.390, P = 0.000), and the correlation coefficients between sElastin and total MMP activity also had significant meaning (r = 0.251, P = 0.003). But there were no such correlations within the CAE group (data not provided).

Fig. 4
figure 4

Correlation between sElastin and total elastase activity and total MMP activity. Pearson’s partial correlation between total elastase activity and sElastin: r = 0.390, P = 0.000; Pearson’s partial correlation between total MMP activity and sElastin: r = 0.251, P = 0.003

Full size image

Discussion

This study mainly evaluated the ECM turnover in CAE populations. CAD group was enrolled in this study because most of the CAE patients have obstructive coronary artery disease [4], and 90 % CAE subjects were associated with CAD in this study. As the baseline characteristics showed, the diastolic blood pressure, body mass index, and blood monocyte number were higher in the CAE and CAD groups. These findings are similar to those observed in other CAE studied by previous researchers [23–25], and are considered characteristics of CAE. The main findings in the present study were as follows: (1) Elastin fibre degradation was significantly increased in the CAE patients according to elevated circulating sElastin. (2) The main change in CAE was overproduction of type I collagen and the excessive degradation of type III collagen. Thus, the ratio of type I collagen to type III collagen was increased in CAE patients. (3) Elastases and MMPs appear to be associated with the elastin fibre degradation. Total elastase activity was significantly higher in CAE group and total MMP activity has a higher tendency in CAE group. And there were considerable correlations between sElastin and total elastase activity, and between sElastin and total MMP activity. There were no such correlations within the CAE group. We speculate that this is possibly due to the limited sample size in the study.

In native vessels, the ECM is composed mostly of elastin fibres, type I and III collagens and proteoglycans. These proteins play important roles in maintaining vessel structure and functions [26, 27]. Elastin fibre, the dominant ECM protein in the coronary wall [12], is composed of soluble elastin subunits, which are synthesised by vascular smooth muscle cells (VSMCs) in the media in response to mechanical and other stimuli. The elastin fibres mainly provide elasticity and tensile strength to the vessels [13]. As elastin is not produced in adults [11], the destruction of elastin fibres is irreversible and may lead to serious damage of ECM integrity [13] and even coronary ectasia. Thus, the increased sElastin further indicates the over degradation of coronary ECM in CAE patients, and could be a promising candidate marker for CAE. The sElastin and its fragments also could promote the disease by: (1) inducing chemotactic attraction of monocytes or macrophages, (2) triggering the production of free radicals and MMPs, (3) promoting the oxidation of LDL, which is then taken up in the vessel wall intima by phagocytic cells, and (4) altering NO• production [10, 12].

Type I and III collagens are the major fibrillar collagen components in vessels, representing 60 and 30 % of vascular collagens, respectively [26]. Changes in the ratio and volume of the two collagens have been implicated in functional properties of the vessels. A higher ratio of collagen type III to I is associated with more compliant tissue, whereas a lower ratio of collagen type III to I results in stiffer, less compliant tissue [14]. In the current study, the ratio of type I to type III collagen was increased in the CAE patients. This type of collagen transition implies that CAE patients have stiffer coronary vessels. A number of studies found evidence of hardening of the arteries in CAE populations, such as an increased coronary artery calcification score [28], higher carotid-femoral pulse wave velocity [29] and a thicker carotid intima media [30]. Thus, coronary artery stiffness seems to be the outcome of an imbalance in collagen metabolism. This type of collagen transition might be a compensatory to prohibit coronary over-dilation.

The total collagen volume appeared to be decreased by comparing the type I collagen synthesis and type III collagen degradation in this study according to the following algorithm: total volume of new synthesis type I collage = serum PICP concentration × total serum volume/molecular mass of PICP mass × molecular mass of type I collagen. Type I collagen is composed of two identical α1 (I) chains (885 aa protein) and one α2 (I) chain (1366 aa protein). PICP was 4.47 ± 2.80 ng/ml in the serum of the CAE patients, had a formula weight of 100,000 g/mol and was cut off from type I procollagen (3136 aa protein) in a ratio of 1:1. The total amount of degraded type III collage can be calculated with a similar equation. Collagen type III is composed of three identical α1 (III) chains (1466 aa protein). PIIINP was 2.20 ± 1.26 ng/ml in the CAE group, had a formula weight of 42,000 g/mol and was cleaved from type III collagen (4398 aa protein) in a similar ratio of 1:1. Thus, the calculated ratio of type I collagen synthesis to type III collagen degradation was 0.608. As a result, there was 0.608 g of type I collagen synthesis to each 1 g of type III collagen degradation. Therefore, the total collagen volume was decreased in the CAE group, with CAE patients having more fragile vessels, and this change may facilitate the ectasia of coronary artery.

In addition to evaluating the turnover of ECM within the CAE group, we also attempted to identify the cause of the degradation of elastin fibres and the transition of type III to type I collagen in ECM metabolism. We focused on related proteases not only because of their indisputable importance as mediators of ECM remodelling but also because previous researches detected changes in the activity of these enzymes in autopsy specimens and serum samples of CAE patients [7–9]. Elastase-type enzymes have been found in all four classes of proteinases, serine, thiol and aspartic proteinases, as well as in several members of the MMP family [11, 26]. Collagenolytic enzymes were mainly MMPs, including at least 26 members [26, 31]. In the present study, the increased activity of both the enzymes could be responsible to some degree, for the degradation of elastin fibres and even the degradation of collagen type III. Proteolytic enzymes seem to play essential roles in the ECM destruction process of coronary media in CAE patients and to be an important aspect of this disease.

In summary, according to results from the detection of circulating markers, the main changes in the ECM metabolism of CAE patients were excessive degradation of elastin fibres, overproduction of type I collagen, increased degradation of type III collagen and a reduction in the total collagen volume. These alterations might contribute to coronary ectasia and stiffness. Proteolytic enzymes possibly took some responsibilities for this kind of abnormal ECM metabolism. Although ECM and proteolytic enzymes were ubiquitous in human organs and tissues, the three groups had distinctive features in the coronary walls, and they were balanced in age, sex and other baseline characteristics. At the same time subjects associated with diseases possibly relating to abnormal ECM metabolism were not included in this study. Thus, those circulating markers could reflect the changes in coronary to some degrees. Further, serum examination was very important in clinical practices due to its convenience and minimal invasion, and human coronary artery autopsy was not practicable, thus optimising the application conditions of those circulating markers (such as combining those markers with untypical chest pain which was the most often symptom of CAE [3] ), and giving more comprehensive explanations for the results would be very helpful for evaluating, diagnosing and even monitoring the disease of coronary ectasia.

Limitations

It would have been better to have divided the CAE group into two sub-groups, CAE with CAD, and CAE without other diseases (isolated CAE). However, the incidence of isolated CAE is rare. Thus, we were unable to enroll sufficient numbers of subjects with balanced baseline characteristics and divide the study groups into subgroups with isolated CAE in this study.